I'm writing a code for identifying the faults in image , Request you to suggest the best way to identifying the faults. Attached an image for reference. Thanks

This is the image which I'm using, Left side shows the original image and the region which I need to extract is shown in right side.

[Walter Roberson]

https://www.mathworks.com/matlabcentral/answers/303635-hi-everyone-i-need-some-help-i-m-working-on-glcm-texture-features-i-want-to-know-which-classifier-gi#answer_235285

Answer 1

[Image Analyst]

I'd create a template of a "good" region. Then use normxcorr2() to do normalized cross correlation on it. Examine the output image for values in a certain range.